Choosing Text-to-Speech or Wave Files

Speech generated by some TTS engines can sound a bit unnatural and may be difficult to understand—many people prefer the natural-sounding voice provided by a wave audio (.wav) file. For this reason, most applications use wave files whenever possible and minimize the use of TTS. Applications typically use TTS to render information that cannot be determined beforehand, such as account balances in an online banking application.

The WTEApplication object includes a configuration property called RenderTTS that enables or disables TTS for a Web telephony application. When disabled, your application cannot render any TTS over the telephone. Disabling TTS is useful if you use wave files throughout your application and have no need for TTS. Alternatively, you might want to use only TTS and not supply any wave files for the application. In this case, of course, you would enable the RenderTTS property.

You can change the setting of the RenderTTS property at any point in your application by using a script. You might start your application with a greeting recorded as a wave file, then enable TTS when your application needs to render the person's checking account balance.

The following example enables TTS by setting the RenderTTS property to TRUE:

window.external.configuration.RenderTTS=true